Upload
hoangtu
View
223
Download
0
Embed Size (px)
Citation preview
Technical Report
FabricPool Best Practices ONTAP 9.4 John Lantz, NetApp
July 2018 | TR-4598
Abstract
This technical report describes best practices for the NetApp® ONTAP® software component
FabricPool. The capabilities, requirements, implementation, and best practices of this software
are covered.
2 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
TABLE OF CONTENTS
Overview ...................................................................................................................................................... 3
1 Primary Use Cases ............................................................................................................................... 4
1.1 Reclaim Space on Primary Storage (Snapshot-Only, Auto) ............................................................................5
1.2 Shrink Secondary Storage Footprint (Backup) ................................................................................................7
1.3 Move Entire Volumes to the Cloud (Backup plus Volume Move) ....................................................................8
2 Requirements ........................................................................................................................................ 9
2.1 Platforms .........................................................................................................................................................9
2.2 Intercluster LIFs ..............................................................................................................................................9
2.3 Volumes ..........................................................................................................................................................9
2.4 FabricPool License ....................................................................................................................................... 10
2.5 Certificate Authority Certification ................................................................................................................... 13
3 Architecture ......................................................................................................................................... 14
3.1 Block Temperature ........................................................................................................................................ 14
3.2 Data Movement ............................................................................................................................................. 14
3.3 Object Storage .............................................................................................................................................. 18
4 Configuration ...................................................................................................................................... 20
4.1 Create a Bucket/Container ............................................................................................................................ 20
4.2 Add an External Capacity Tier to ONTAP ..................................................................................................... 25
4.3 Attach an External Capacity Tier to an Aggregate ........................................................................................ 27
4.4 Set Volume Tiering Policies .......................................................................................................................... 29
4.5 Security ......................................................................................................................................................... 31
5 Interoperability .................................................................................................................................... 31
6 Performance ........................................................................................................................................ 33
6.1 Sizing ............................................................................................................................................................ 33
6.2 Connectivity .................................................................................................................................................. 34
6.3 Capacity ........................................................................................................................................................ 35
6.4 NetApp Private Storage for AWS .................................................................................................................. 39
Where to Find Additional Information .................................................................................................... 40
Version History ......................................................................................................................................... 41
Contact Us ................................................................................................................................................. 41
3 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Overview
FabricPool, first available in ONTAP 9.2, is a NetApp Data Fabric technology that enables automated
tiering of data to low-cost object storage tiers either on or off premises.
Unlike manual tiering solutions, FabricPool reduces total cost of ownership by automating the tiering of
data to lower the cost of storage. It delivers the benefits of cloud economics by tiering to public clouds
such as Amazon S3 and Microsoft Azure Blob Storage as well as to private clouds created with NetApp
StorageGRID® Webscale.
FabricPool is transparent to applications and allows enterprises to take advantage of cloud economics
without sacrificing performance or having to rearchitect solutions to leverage storage efficiency.
• ONTAP supports FabricPool on AFF systems and all-SSD aggregates on FAS systems.
• ONTAP Select supports FabricPool. NetApp recommends using all-SSD FabricPool aggregates.
• Cloud Volumes ONTAP for AWS supports FabricPool on gp2, io1, and st1 EBS volumes.
• Cloud Volumes ONTAP for Azure supports
Figure 1) Before and after FabricPool.
4 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
1 Primary Use Cases
The primary purpose of FabricPool is to reduce storage footprints and associated costs. Active data
remains on high-performance SSDs; inactive data is tiered to low-cost object storage while preserving
ONTAP data efficiencies.
FabricPool has three primary use cases in ONTAP 9.4:
• Reclaiming space on primary storage
• Shrinking secondary storage
• Moving volumes to external capacity tiers
Although FabricPool can significantly reduce storage footprints, it is not a backup solution. WAFL® (Write
Anywhere File Layout) metadata always stays on the performance tier. If a catastrophic disaster destroys
the performance tier, a new environment cannot be created using the data on the capacity tier because it
contains no WAFL metadata.
For complete protection that includes WAFL metadata, consider using existing ONTAP technologies like
SnapMirror and SnapVault. FabricPool can be used to reduce costs by tiering data on secondary storage
as well.
5 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
1.1 Reclaim Space on Primary Storage (Snapshot-Only, Auto)
Snapshot-Only Tiering Policy
Snapshot copies can frequently consume more than 10% of a typical storage environment. Although
essential for data protection and disaster recovery, these point-in-time copies are rarely used and are an
inefficient use of high-performance SSDs.
Snapshot-Only, the default volume tiering policy for FabricPool, is an easy way to reclaim storage space
on SSDs. When configured to use this policy, cold Snapshot blocks in the volume that are not shared with
the active file system are moved to the capacity tier. If read, cold data blocks on the capacity tier become
hot and are moved to the performance tier.
Note: The FabricPool Snapshot-Only volume tiering policy reduces the amount of storage used by Snapshot copies on SSDs. It does not increase the maximum number of Snapshot copies allowed by ONTAP, which remains 1,023.
Figure 2) Reclaiming space with the Snapshot-Only volume tiering policy.
6 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Auto Tiering Policy
The majority of inactive (cold) data in storage environments is associated with unstructured data,
accounting for more than 50% of total storage capacity in many storage environments.
Infrequently accessed data associated with productivity software, completed projects, and old datasets is
an inefficient use of high-performance SSDs, and tiering this data to a low-cost object store is an easy
way to reclaim existing SSD capacity and reduce the amount of required SSD capacity moving forward.
First available in ONTAP 9.4, the Auto volume tiering policy moves all cold blocks in the volume, not just
blocks associated with Snapshot copies, to the capacity tier.
If read by random reads, cold data blocks on the capacity tier become hot and are moved to the
performance tier. If read by sequential reads such as those associated with index and antivirus scans,
cold data blocks on the capacity tier stay cold and are not written to the performance tier.
Figure 3) Reclaiming space with the Auto volume tiering policy.
7 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
1.2 Shrink Secondary Storage Footprint (Backup)
Secondary data (data protection volumes that are NetApp SnapMirror® [disaster recovery] or NetApp
SnapVault® [backup] destination targets) is frequently stored on secondary clusters that share a 1:1 or
greater ratio with the primary data that they are protecting (one baseline copy and multiple Snapshot
copies). For large datasets, this approach can be prohibitively expensive, forcing users to make
expensive decisions about the data they need to protect.
Like Snapshot copies, backups are infrequently used, are an inefficient use of high-performance SSDs,
and are expensive for large datasets even when using HDDs. FabricPool’s Backup volume tiering policy
changes this paradigm.
Instead of 1:1 primary-to-backup ratios, the FabricPool Backup policy allows users to significantly reduce
the number of disk shelves on their secondary clusters, tiering the majority of the backup data to low-cost
object stores. WAFL metadata remains on the secondary cluster’s performance tier.
If read, cold data blocks in volumes using the Backup policy are not written to the performance tier. This
reduces the need for high-capacity secondary storage performance tiers.
Figure 4) Backup use case.
8 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
1.3 Move Entire Volumes to the Cloud (Backup plus Volume Move)
In addition to Snapshot copies and backups, one of the most common uses of FabricPool is to move data
to an external capacity tier. Completed projects, legacy reports, or historical records—any dataset that
must be retained but is unlikely to be frequently read—are ideal candidates to be moved to low-cost
object storage.
Moving entire volumes is accomplished by setting the Backup volume tiering policy on a volume when
initiating a volume move. A volume move is the only time when a backup tiering policy can be set on a
volume that is not a data protection destination target.
During the volume move, data (excluding WAFL metadata) is immediately moved to the external capacity
tier attached to the destination aggregate. After the volume move has completed, the tiering policy on the
volume automatically reverts from Backup to Auto.
If read, cold data blocks on the capacity tier become hot and are moved to the performance tier.
Figure 5) Volume move use case.
9 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
2 Requirements
FabricPool requires ONTAP 9.2 or later and the use of SSD aggregates on any of the platforms listed in
the next section. Additional FabricPool requirements depend on the capacity tier being attached.
Although installation and use of certificate authority (CA) certificates are recommended best practices,
starting in ONTAP 9.4, installation of CA certificates is not required for StorageGRID Webscale.
2.1 Platforms
FabricPool is supported on a variety of platforms:
• AFF
− A800
− A700S, A700
− A300
− A220, A200
− AFF8080, AFF8060, AFF8040
• FAS*
− FAS9000
− FAS8200
− FAS8080, FAS8060, FAS8040
− FAS2750, FAS2720
− FAS2650, FAS2620
Note: Only SSD aggregates can use FabricPool.
• Cloud Volumes ONTAP for AWS
− Performance tier: gp2, io1, and st1 EBS volumes
− The Auto volume tiering policy is not supported with Standard pricing
• Cloud Volumes ONTAP for Azure
• Capacity tier
− Amazon S3 (Standard and Standard - Infrequent Access)
− AWS GovCloud S3*
− Microsoft Azure Blob Storage (Hot and Cool)
− StorageGRID Webscale 10.3+
*FabricPool does not currently support FIPS 140-2–compliant AWS GovCloud endpoints.
2.2 Intercluster LIFs
Cluster high-availability (HA) pairs that use FabricPool require two intercluster LIFs to communicate with
the external capacity tier. NetApp recommends creating an intercluster LIF on additional HA pairs to
seamlessly attach external capacity tiers to aggregates on those nodes as well.
Note: Disabling or deleting an intercluster LIF interrupts communication to the external capacity tier.
2.3 Volumes
FabricPool cannot attach an external capacity tier to an aggregate that contains volumes using a space
guarantee other than None (for example, Volume).
volume modify -space-guarantee none
10 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Setting the space-guarantee none parameter assures thin provisioning of the volume. The amount of
space consumed by volumes with this guarantee type grows as data is added instead of being
determined by the initial volume size. This approach is essential for FabricPool because the volume must
support external capacity tier data that becomes hot and is brought back to the performance tier.
2.4 FabricPool License
FabricPool requires a capacity-based license when attaching third-party object storage providers (such as
Amazon S3) as capacity tiers for AFF and FAS hybrid flash systems. A FabricPool license is not required
when using StorageGRID Webscale as the capacity tier or when using Amazon S3 or Microsoft Azure
Blob Storage as the capacity tier for Cloud Volumes for ONTAP.
Tiering to the capacity tier stops when the amount of data (used capacity) stored on the third capacity tier
reaches the licensed capacity. Additional data, including SnapMirror copies to volumes using the backup
tiering policy, cannot be tiered until the license capacity is increased. Although tiering stops, data remains
accessible from the capacity tier. Additional cold data remains on SSDs until the licensed capacity is
increased.
A free 10TB capacity FabricPool license comes with the purchase of any new ONTAP 9.2+ cluster, but
because FabricPool is a feature introduced in ONTAP 9.2, existing clusters do not have a FabricPool
license by default. FabricPool licenses (and additional capacity for existing licenses) can be purchased in
1TB increments.
Note: FabricPool licenses are clusterwide. Have your UUID available when purchasing a license (cluster identify show). For additional licensing information, refer to the NetApp Knowledgebase.
11 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Installation
After you acquire a NetApp license file (NLF), you can install it by using OnCommand® System Manager.
To do so, complete the following steps:
1. Click Configurations.
2. Click Licenses.
3. Click Add.
4. Click Choose Files to browse and select a file.
5. Click Add.
Figure 6) FabricPool license.
12 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
License Capacity
Licensed capacity can be viewed using the ONTAP CLI as well as OnCommand System Manager.
To see the licensed capacity, run the following command in the ONTAP CLI:
system license show-status
In OnCommand System Manager, complete the following steps:
1. Click Configurations.
2. Click Licenses.
3. Click the Details tab.
Maximum capacity and current capacity are listed on the FabricPool License row.
13 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
2.5 Certificate Authority Certification
When FabricPool uses StorageGRID Webscale as a capacity tier, it must use a Transport Layer Security
(TLS) connection. Using Amazon S3 without TLS configuration is supported, but not recommended.
FabricPool requires that CA certificates use the same fully qualified domain name (FQDN) as the capacity
tier server with which they are associated, but the default StorageGRID Webscale CA certificates use a
common name (CN) that isn’t based on the server’s FQDN. This approach causes certificate-based errors
that prohibit StorageGRID Webscale from being attached to aggregates.
Errors might include the following examples:
• “Unable to add external capacity tier. Cannot verify the certificate provided by the object store server. The certificates might not be installed on the cluster. Do you want to add the certificate now?”
• “Cannot verify the certificate provided by the object store server.”
To avoid these errors and successfully attach StorageGRID Webscale as a capacity tier, you must
replace the certificates in the grid with certificates that use the correct FQDN.
Although self-signed certificates can be used, using signed certificates from a third-party certificate
authority is the recommended best practice.
Note: Starting in ONTAP 9.4, CA certificates are no longer required. Using signed certificates from a third-party certificate authority remains the recommended best practice.
Installation
To install CA certificates in ONTAP, complete the following steps:
1. Copy the StorageGRID Webscale certificates.
2. Paste the certificates into ONTAP.
Copy Certificates from StorageGRID Webscale
To copy certificates by using StorageGRID Webscale, complete the following steps:
1. Click Configuration.
2. Click Grid Options.
3. From the Grid Options menu, click Overview.
4. Under API Server Certificates, expand CA Certificate.
5. Copy the CA certificate, including the following text:
-----BEGIN CERTIFICATE----- and -----END CERTIFICATE-----
To copy certificates by using OpenSSL, complete the following steps:
1. Run the following command:
openssl s_client -connect <FQDN> -showcerts
2. Copy the CA certificate, including the following text:
-----BEGIN CERTIFICATE----- and -----END CERTIFICATE-----
Paste Certificates to ONTAP
To paste certificates to ONTAP, run the following command:
security certificate install -type server-ca
14 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
3 Architecture
FabricPool works by associating an external object store (such as Amazon S3, Microsoft Azure Blob
Storage, or StorageGRID Webscale) with an aggregate in ONTAP, creating a composite aggregate: a
FabricPool.
Volumes inside the composite aggregate can then take advantage of the FabricPool by keeping active
(hot) data on performative SSDs (the performance tier) and tiering inactivate (cold) data to the external
object store (the capacity tier).
Although only a basic level of understanding is necessary to configure and use FabricPool, understanding
how FabricPool determines block temperature, migrates data, and creates data objects is extremely
useful when architecting storage solutions.
3.1 Block Temperature
When a block is written to an SSD, it is assigned a temperature value indicating that it is hot. Over time, a
background cooling scan cools blocks, making hot blocks warm and eventually turning blocks cold if they
have not been read. Assuming no activity, a block becomes cold based on the time set by the tiering-
minimum-cooling-days setting.
Note: The Backup volume tiering policy is an exception to this rule. Blocks in volumes using the Backup tiering policy are immediately identified as cold and marked for tiering.
Note: Cloud Volumes for ONTAP must be active for the background cooling scan to work. Turning off a Cloud Volumes for ONTAP instance when it is not in use might reduce compute costs, but it also prevents the cooling scan from running.
3.2 Data Movement
After a block has been identified as cold, it is marked for tiering. During this time, a background tiering
scan looks for cold blocks. When enough 4KB blocks from the same volume have been collected, they
are concatenated into a 4MB object and moved to the capacity tier based on the volume tiering policy.
Tiering to the capacity tier only happens if the performance tier aggregate is >50% full. There is little
reason to tier cold data to an external capacity tier if the performance tier is being underutilized. This 50%
performance tier utilization requirement cannot be modified by storage administrators.
If tiering is disabled on a volume (the volume tiering policy is set to None), blocks on the capacity tier
remain there until they become hot or a volume move is performed. If read, a block immediately becomes
hot and is moved back to the performance tier. However, there are two exceptions to this rule:
• The performance tier is more than 70% capacity. If this situation occurs, cold data is read directly from the external capacity tier without migrating to the performance tier.
• When performing a SnapMirror volume restore, if both source and destination volumes use the Snapshot-Only policy and the Snapshot block is being read from the external capacity tier, then FabricPool knows the blocks are cold, and they are restored directly to the external capacity tier.
15 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
SnapMirror Behavior
Movement of data from the external capacity tier to the performance tier can take place any time a block
is read.
Table 1) SnapMirror behavior.
Source Volume Tiering Policy
Destination Volume Tiering Policy
Write Location
Auto Auto Performance > Performance Capacity > Capacity
Auto Snapshot-Only Performance
Auto Backup Capacity
Auto None Performance
Snapshot-Only Auto Performance > Performance Capacity > Capacity
Snapshot-Only Snapshot-Only Performance > Performance Capacity > Capacity
Snapshot-Only Backup Capacity
Snapshot-Only None Performance
Backup Auto Performance
Backup Snapshot-Only Performance
Backup Backup Capacity
Backup None Performance
None Auto Performance
None Snapshot-Only Performance
None Backup Capacity
None None Performance
Volume Move
Volume move (vol move) is the way that ONTAP moves a volume nondisruptively from one aggregate
(source) to another (destination). Volume moves can be performed for a variety of reasons, although the
most common reasons are generally hardware lifecycle management, cluster expansion, and load
balancing.
It is important to understand how volume move works with FabricPool because the changes that take
place at both the aggregate (the attached external capacity tier) and the volume (volume tiering policies)
levels can have a major impact on functionality.
16 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Destination Aggregates
If a volume move’s destination aggregate does not have an attached capacity tier, data on the source
volume that is stored on the external capacity tier is migrated to the performance tier on the destination
aggregate.
If a volume move’s destination aggregate has an attached capacity tier, data on the source volume that is
stored on the external capacity tier is first migrated to the performance tier on the destination aggregate. It
is then migrated to the external capacity tier on the destination aggregate if this approach is appropriate
for the volume’s tiering policy. Migrating data to the performance tier first improves the performance of the
volume move and reduces cutover time.
If a volume tiering policy is not specified when performing a volume move, the destination volume uses
the tiering policy of the source volume. If a different tiering policy is specified when performing the volume
move, the destination volume is created with the specified tiering policy.
Snapshot-Only
If -tiering-policy snapshot-only is specified during the volume move, data movement is
variable, but data moves to the destination aggregate’s performance tier first.
If both source and destination volumes use the Snapshot-Only policy, and the Snapshot block is being
read from the source aggregate’s external capacity tier, then FabricPool knows the Snapshot blocks are
cold and moves the cold blocks to the destination aggregate’s external capacity tier.
vol move start -vserver <name> -volume <name> -destination-aggregate <name> -tiering-policy
snapshot-only
Auto
If -tiering-policy auto is specified during the volume move, data movement is variable, but all data
moves to the destination aggregate’s performance tier first.
If the source volume uses the Auto, None, or Snapshot-Only policy, blocks are moved to the same tier
that they existed on prior to the move. If the source volume uses the Backup policy all data is moved to
the performance tier.
vol move start -vserver <name> -volume <name> -destination-aggregate <name> -tiering-policy auto
Backup
If -tiering-policy backup is specified during the volume move, data is immediately identified as
cold and migrated to the destination aggregate’s external capacity tier. There is no need to wait 48 hours
for blocks in the volume to become cold. Metadata is always stored on the performance tier.
vol move start -vserver <name> -volume <name> -destination-aggregate <name> -tiering-policy
backup
None
If -tiering-policy none is specified during the volume move, data is migrated to the destination
aggregate’s performance tier.
vol move start -vserver <name> -volume <name> -destination-aggregate <name> -tiering-policy none
17 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
OnCommand System Manager
To perform a volume move with OnCommand System Manager, complete the following steps:
1. Navigate to the Volumes page.
2. Select the volume you want to move.
3. Click Actions.
4. Click Move.
5. Select a destination aggregate.
6. Select the Backup tiering policy.
7. Click Move.
18 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
ONTAP CLI
To perform a volume move using the ONTAP CLI, run the following command:
vol move start -vserver <name> -volume <name> -destination-aggregate <name> -tiering-policy
backup
Note: A volume move using the ONTAP CLI or the Volume Move wizard in OnCommand System Manager is the only time when a Backup tiering policy can be set on a non–data protection volume.
After the volume move has completed, the volume tiering policy on the volume automatically reverts from
Backup to Snapshot-Only if the volume is not a data protection volume. Data remains on the capacity tier
unless it is made hot and returns to the performance tier. If both the source and destination volumes are
data protection volumes, the volume tiering policy remains set to Backup.
3.3 Object Storage
Object storage is a storage architecture that manages data as objects, as opposed to other storage
architectures such as file or block storage. Objects are kept inside a single container (such as a bucket)
and are not nested as files inside a folder inside other folders.
Although object storage is generally less performative than file or block storage, it is significantly more
scalable. An ONTAP performance tier currently has a maximum volume size of 100TB and a maximum
aggregate size of 800TB. Object stores have no such limits, and buckets with petabytes of data in them
are not uncommon.
FabricPool Objects
FabricPool works at the block level, cooling blocks, concatenating them into objects, and migrating those
objects to an external capacity tier. Each FabricPool object is 4MB and composed of 1,024 4KB blocks.
The object size is fixed at 4MB based on performance recommendations from Amazon and NetApp and
cannot be changed. If cold blocks are read and made hot, only the requested 4KB blocks in the 4MB
object are fetched. Neither the entire object nor the entire file is migrated back. Only the necessary blocks
are migrated.
Note: If ONTAP detects sequential readaheads, it requests blocks from the external capacity tier before they are read in order to improve performance.
FabricPool Object Stores
FabricPool currently supports Amazon S3 (Standard and Standard - IA), Microsoft Azure Blob Storage
(Hot and Cool), and StorageGRID Webscale 10.3+ object stores as external capacity tiers.
More than one type of object store can be connected to a cluster, but only one type of object store can be
attached to each aggregate. For example, one aggregate can use StorageGRID Webscale, and another
aggregate can use Amazon S3, but one aggregate cannot be attached to both.
Note: FabricPool does not support information lifecycle management (ILM) policies applied to object store buckets. ILM typically includes various movement and deletion policies that would be disruptive to FabricPool capacity tier data. FabricPool has no knowledge of ILM policies or configurations set on external object stores, and misconfiguration of ILM policies can result in data loss.
Object Deletion and Defragmentation
FabricPool does not delete blocks from the object store. Instead, it deletes entire objects after a certain
percentage of the blocks in the object have been deleted or are no longer referenced by the performance
tier.
19 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
For example, there are 1,024 4KB blocks in a 4MB object tiered to Amazon S3. Defragmentation and
deletion do not occur until less than 205 4KB blocks (20% of 1,024) are being referenced by ONTAP.
When enough (1,024) blocks have zero references, their original 4MB objects are deleted, and a new
object is created.
This percentage, the unreclaimed space threshold, can be customized, but is set to different default levels
for different object stores. The default settings are as follows:
• 15% Azure Blob Storage
• 20% Amazon S3
• 40% StorageGRID Webscale
Unreclaimed Space Threshold
Object defragmentation reduces the amount of physical capacity used by the capacity tier at the expense
of additional object store resources (reads and writes).
To avoid additional expenses, consider reducing the default unreclaimed space thresholds when using
object store pricing schemes that reduce the cost of storage but increase the cost of reads. Examples
include Amazon’s Standard - IA and Azure Blob Storage’s cool.
For example, tiering a volume of 10-year-old projects that has been saved for legal reasons might be less
expensive when using a pricing scheme such as Standard - IA or cool than it would be when using
standard pricing schemes. Although reads are more expensive for such a volume, including reads
required by object defragmentation, they are unlikely to occur frequently here.
storage aggregate object-store modify –aggregate <name> -object-store-name <name> –unreclaimed-
space-threshold <%> (0%-99%)
(Advanced privilege level required.)
Note: Prior to ONTAP 9.4, object deletion effectively takes place at 0% free.
ONTAP Storage Efficiencies
Storage efficiencies such as compression, deduplication, and compaction are preserved when moving
data to the capacity tier. Aggregate inline deduplication is supported on the performance tier, but these
efficiencies are not carried over to objects stored on the external capacity tier.
Note: Third-party, server-side deduplication has not been qualified by NetApp.
20 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
4 Configuration
After FabricPool’s basic requirements have been met, attaching Amazon S3, Microsoft Azure Blob
Storage, or StorageGRID Webscale to an aggregate in ONTAP requires the following four steps:
1. Create a bucket.
2. Add an external capacity tier to ONTAP.
3. Attach an external capacity tier to an aggregate.
4. Set volume tiering policies.
4.1 Create a Bucket/Container
Buckets are object store containers that hold data. You must provide the name and location of the bucket
where data is stored before it can be added to an aggregate as an external capacity tier.
Creating a bucket in StorageGRID Webscale is not as intuitive because it must be done using third-party
tools. StorageGRID Webscale buckets cannot be created using the StorageGRID Webscale grid
management interface, OnCommand System Manager, OnCommand Unified Manager, or ONTAP.
FabricPool supports the attachment of one bucket per aggregate. A single bucket can be attached to a
single aggregate, and a single bucket can be attached to multiple aggregates. However, a single
aggregate cannot be attached to multiple buckets. Although a single bucket can be attached to multiple
aggregates in a cluster, NetApp does not recommend attaching a single bucket to aggregates in multiple
clusters.
Note: Consider how bucket-to-aggregate relationships might affect performance when planning storage architectures. Many object store providers set a maximum number of supported IOPS at the bucket/container level. Environments that require maximum performance should use multiple buckets to reduce the possibility that object store IOPS limitations affect performance across multiple FabricPool aggregates. Attaching a single bucket/container to all FabricPool aggregates in a cluster might be more beneficial to environments that value manageability over external capacity tier performance.
Figure 7) Bucket-to-aggregate relationships.
21 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Amazon S3
To create a bucket in Amazon S3, complete the following steps:
1. Sign in to the AWS management console and open the S3 console:
https://console.aws.amazon.com/s3
2. Click Create bucket.
3. Enter a unique DNS-compliant bucket name.
4. Select a region close to your data center.
5. Click Create.
Additional Settings
FabricPool does not support ILM policies applied to object store buckets. ILM typically includes various
movement and deletion policies that could be disruptive to FabricPool capacity tier data. FabricPool has
no knowledge of ILM policies or configurations set on Amazon S3, and misconfiguration of ILM policies
can result in data loss.
22 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Microsoft Azure Blob Storage
To create a container in Microsoft Azure Blob Storage, complete the following steps:
1. Sign in to the Microsoft Azure dashboard:
https://portal.azure.com
2. Click Settings/Containers.
3. Enter a name.
4. Select an access level.
5. Click OK.
Additional Settings
FabricPool does not support ILM policies applied to object store buckets. ILM typically includes various
movement and deletion policies that could be disruptive to FabricPool capacity tier data. FabricPool has
no knowledge of ILM policies or configurations set by Microsoft Azure Blob Storage, and misconfiguration
of ILM policies can result in data loss.
23 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
StorageGRID Webscale
An S3 browser is required to create a bucket in StorageGRID Webscale. Although many S3 browser
options are available, CloudBerry Explorer was used for the following instructions. To create a bucket in
StorageGRID Webscale, complete the following steps:
1. Click File.
2. Select New S3 Compatible Account.
3. Select S3 Compatible Account.
4. Enter the display name, service point, access key, and secret key associated with your StorageGRID Webscale account.
5. Click Test Connection.
6. Click OK.
7. Select a storage account (StorageGRID in this example) as the destination source.
8. Click the New Bucket icon.
24 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
9. Name the bucket.
10. Click OK.
Additional Settings
Other than StorageGRID Webscale data replication and erasure coding, FabricPool does not support ILM
policies applied to object store buckets. ILM typically includes various movement and deletion policies
that could be disruptive to FabricPool capacity tier data. FabricPool has no knowledge of ILM policies or
configurations set on StorageGRID Webscale, and misconfiguration of ILM policies can result in data
loss.
Note: Data replication protects objects from loss by making copies of object data and storing those copies on multiple storage or archive nodes. ILM rules dictate the number, location, and retention time of copies.
OnCommand Cloud Manager
OnCommand Cloud Manager automatically creates an S3 bucket in Cloud Volumes for ONTAP when S3
tiering has been enabled by selecting the Use S3 as a Capacity Tier for Cold Data checkbox.
Figure 8) OnCommand Cloud Manager S3 Tiering checkbox.
25 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
4.2 Add an External Capacity Tier to ONTAP
Before an object store can be attached to an aggregate, it must be added to and identified by ONTAP.
This task can be completed using either OnCommand System Manager or the ONTAP CLI.
FabricPool currently supports Amazon S3, Microsoft Azure Blob Storage, and StorageGRID Webscale
10.3+ object stores as external capacity tiers.
You need the following information:
• Server name (FQDN) (for example, s3.amazonaws.com)
• Access key ID
• Secret key
• Container name (bucket name)
OnCommand System Manager
To add an external capacity tier using OnCommand System Manager, complete the following steps:
1. Launch OnCommand System Manager.
2. Click Storage.
3. Click Aggregates & Disks.
4. Click External Capacity Tiers.
5. Select an object store provider.
6. Complete the text fields as required for your object store provider.
Note: Enter the object store’s bucket/container name in the Container Name field.
7. Click Save and Attach Aggregates.
26 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
ONTAP CLI
To add an external capacity tier using the ONTAP CLI, enter the following commands:
object-store config create
-object-store-name <name>
-provider-type <AWS/Azure_Cloud/SGWS>
-port <443/8082> (AWS&Azure_Cloud/SGWS)
-server <name>
-container-name <bucket-name>
-access-key <string>
-secret-password <string>
-ssl-enabled true
-ipspace default
Cloud Volumes for ONTAP
Virtual private cloud (VPC) endpoints to the Amazon S3 service must be enabled before tiering can take
place in Cloud Volumes for ONTAP. This approach ensures secure, encrypted communication to and
from your Amazon S3 buckets over the AWS network.
When you create the VPC endpoint, be sure to select the region, VPC, and VPC route table that
correspond to the Cloud Volumes for ONTAP instance.
Certificate Authority Certificate Validation
As a recommended best practice, CA certificates associated with private cloud object stores, such as
StorageGRID Webscale, should be installed on ONTAP before attaching them to aggregates as an
external capacity tier. Using CA certificates creates a trusted relationship between ONTAP and the object
store and helps to secure access to management interfaces, gateway nodes, and storage.
A failure to install a CA certificate results in an error unless certificate validation is turned off. Turning off
certificate validation is not recommended, but it is possible starting in ONTAP 9.4.
OnCommand System Manager
CA certificate validation can be turned off when adding an external capacity tier using OnCommand
System Manager. To do so, complete the following steps:
1. Launch OnCommand System Manager.
2. Click Storage.
3. Click Aggregates & Disks.
4. Click External Capacity Tiers.
5. Select an object store provider.
6. Complete the text fields as required for your object store provider.
7. Click the Object Store Certificate button to turn it off.
8. Click Save and Attach Aggregates.
27 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Figure 9) Object store certificate.
ONTAP CLI
CA certificate validation can be turned off when adding an external capacity tier using the ONTAP CLI. To
do so, run the following commands:
object-store config create
-object-store-name <name>
-provider-type <AWS/Azure_Cloud/SGWS>
-port <443/8082> (AWS&Azure_Cloud/SGWS)
-server <name>
-container-name <bucket-name>
-access-key <string>
-secret-password <string>
-ssl-enabled true
-ipspace default
-is-certificate-validation-enabled false
4.3 Attach an External Capacity Tier to an Aggregate
After an object store has been added to and identified by ONTAP, it must be attached to an aggregate to
create a FabricPool. This task can be completed using either OnCommand System Manager or the
ONTAP CLI.
More than one type of object store can be connected to a cluster, but only one type of object store can be
attached to each aggregate. For example, one aggregate can use StorageGRID Webscale, and another
aggregate can use Amazon S3, but one aggregate cannot be attached to both.
Note: Attaching an external capacity tier to an aggregate is a permanent action. After being attached, an
external capacity tier cannot be removed.
28 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Thin Provisioning
FabricPool cannot attach an external capacity tier to an aggregate that contains volumes using a space
guarantee other than none (for example, volume). For additional information, refer to FabricPool’s
requirements.
OnCommand System Manager
To attach an external capacity tier to an aggregate using OnCommand System Manager:
1. Launch OnCommand System Manager.
2. Click Applications and Tiers.
3. Click Storage Tiers.
4. Click an aggregate.
5. Click Actions and select Attach External Capacity Tier.
6. Select an external capacity tier.
7. View and update the tiering policies for the volumes on the aggregate (optional). By default, volume tiering policies are set as Snapshot-Only.
8. Click Save.
ONTAP CLI
To attach an external capacity tier using the ONTAP CLI, run the following commands:
storage aggregate object-store attach
-aggregate <name>
-object-store-name <name>
Example:
storage aggregate object-store attach -aggregate aggr1 -object-store-name -aws_fabricpool_bucket
OnCommand Cloud Manager
Select the Use S3 as a Capacity Tier for Cold Data checkbox to attach an external capacity tier using
Cloud Volumes for ONTAP when creating a volume or changing a tier:
Note: Cloud Volumes for ONTAP supports S3 capacity tiers on gp2, io1, and st1 EBS volumes.
29 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
4.4 Set Volume Tiering Policies
All volumes are assigned the Snapshot-Only tiering policy by default. The exception to this rule is that
cloned volumes inherit the policy of their parent volume by default. After volume creation, the volume
tiering policy can be changed using OnCommand System Manager or the ONTAP CLI.
Note: It is important to understand how volume move works with FabricPool because the changes that take place at both the aggregate (the attached external capacity tier) and volume (volume tiering policies) levels can have a major impact on functionality.
FabricPool provides four volume tiering policies, as described in the following sections.
• Snapshot-Only (Default)
− Cold Snapshot blocks in the volume that are not shared with the active file system are moved to the capacity tier. Assuming the aggregate is >50% utilized, it takes approximately two days for inactive Snapshot blocks to become cold. The Snapshot-Only cooling period is adjustable from 2 to 63 days using tiering-minimum-cooling-days.
− When cold blocks in a volume with a tiering policy set to Snapshot-Only are read, they are made hot and written to the performance tier.
• Auto
− All cold blocks in the volume are moved to the capacity tier. Assuming the aggregate is >50% utilized, it takes approximately 31 days for inactive blocks to become cold. The Auto cooling period is adjustable between 2 days and 63 days using tiering-minimum-cooling-days.
− When cold blocks in a volume with a tiering policy set to Auto are read randomly, they are made hot and written to the performance tier.
− When cold blocks in a volume with a tiering policy set to Auto are read sequentially, they stay cold and remain on the capacity tier. They are not written to the performance tier.
Note: The Auto volume tiering policy is not supported on Cloud Volumes for ONTAP Standard.
• Backup
− The backup policy can only be set on data protection volumes (destination targets).
− All data blocks (not including WAFL metadata) in the volume are immediately moved to the capacity tier.
− Unlike volumes using the Snapshot-Only tiering policy, there is no need to wait 48 hours for blocks in a volume using the backup tiering policy to become cold.
− When cold blocks in a volume with a tiering policy set to Backup are read, they stay cold and remain on the capacity tier. They are not written to the performance tier.
Note: If the volume’s data protection relationship is broken, the destination volume is no longer a data protection volume, and the tiering policy reverts to the Auto volume tiering policy.
• None
− Volumes set to use none as their tiering policy do not tier cold data to the capacity tier.
− Setting the tiering policy to none prevents new tiering. Volume data that has previously been moved to the external capacity tier remains in the external capacity tier until it becomes hot and is automatically moved back to the performance tier.
− When cold blocks in a volume with a tiering policy set to none are read, they are made hot and written to the performance tier.
30 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
OnCommand System Manager
To change a volume’s tiering policy by using OnCommand System Manager, complete the following
steps:
1. Launch OnCommand System Manager.
2. Select a volume.
3. Click Actions and select Change Tiering Policy.
4. Select the tiering policy you want to apply to the volume. By default, volume tiering policies are set as Snapshot-Only.
5. Click Save.
ONTAP CLI
To change a volume’s tiering policy using the ONTAP CLI, run the following commands:
volume modify -vserver <svm_name> -volume <volume_name>
-tiering-policy <snapshot-only|auto|backup|none>
Note: The default is Snapshot-Only.
-tiering-minimum-cooling-days <2-63>
(Advanced privilege level required.)
Note: The default is two days for Snapshot-Only and 31 days for Auto.
OnCommand Cloud Manager
Volume tiering policies can be set using OnCommand Cloud Manager when creating new volumes, or by
clicking Change Disk Type & Tiering Policy on existing volumes. OnCommand Cloud Manager sets the
Auto policy on source volumes and the Backup policy on data protection volumes automatically.
Note: The Auto volume tiering policy is not supported on Cloud Volumes for ONTAP Standard. OnCommand Cloud Manager sets the Snapshot-Only policy by default on source volumes in Cloud Volumes for ONTAP Standard.
31 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
4.5 Security
FabricPool maintains AES-256 encryption on the performance tier, on the external capacity tier, and over
the wire when moving data between the tiers.
Performance Tier
FabricPool supports NetApp Storage Encryption (NSE) and NetApp Volume Encryption (NVE). Neither
NSE, nor NVE, are required to use FabricPool aggregates.
Over the Wire
Objects moving between performance and capacity tiers are encrypted by using TLS 1.2. To some
extent, encryption affects connectivity (latency) because object stores must use CPU cycles to decrypt
the data. Communicating with object stores without TLS encryption is supported but is not recommended.
Capacity Tier
All objects are encrypted server-side using AES-256 encryption. Server-side encryption keys are owned
by the respective object store.
5 Interoperability
In general, ONTAP functionality is unchanged on FabricPool aggregates. Although ONTAP must create
and transfer objects and blocks between performance and capacity tiers, data protection, efficiency, and
security are nearly identical to those for standard aggregates in ONTAP. The primary differentiators are
performance and cost, object stores being slower and less expensive.
The exceptions to normal interoperability listed in Table 2 and Table 3 are unique to FabricPool
aggregates.
Table 2) NetApp interoperability.
Focus Supported Not Supported
Capacity tier StorageGRID Webscale 10.3+ HDD aggregates
Data protection SnapMirror
SnapVault
StorageGRID Webscale replication and erasure coding Note: for best performance, use StorageGRID Webscale 11.1+ when enabling erasure coding
7-Mode Data Transition Using SnapMirror®
7-Mode Transition Tool (7MTT)
Dump
MetroCluster™
NDMP (including NDMPcopy check with your backup vendor to determine if NDMPCopy is required)
SMTape
SnapLock®
SVMDR
StorageGRID Webscale ILM policies other than replication and erasure coding
StorageGRID Webscale object versioning
32 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Encryption NetApp Volume Encryption
NetApp Storage Encryption
Server-side encryption (AES-256)
TLS 1.2
–
Storage efficiency Inline deduplication
Inline compression
Compaction
Aggregate inline deduplication (performance tier only)
Storage virtualization
NetApp® FlexArray
Quality of service (QoS)
QoS maximums (floors) QoS minimums (floors)
Additional features Auto Balance Aggregate
Table 3) Third-party interoperability.
Focus Supported Not Supported
Capacity tier Amazon S3 (Standard and Standard - IA)
Microsoft Azure Blob Storage (Hot and Cool)
Amazon Glacier
Azure Archive
Data protection Amazon’s 99.999999999% multi-region durability
ILM policies
Encryption Server-side encryption (AES-256)
TLS 1.2
–
33 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
6 Performance
6.1 Sizing
When considering sizing, keep in mind that the performance tier should be capable of the following tasks:
• Supporting hot data
• Supporting cold data until the tiering scan puts the objects on the capacity tier
• Supporting external capacity tier data that becomes hot and is written back to the performance tier
• Supporting all WAFL metadata associated with the attached external capacity tier
For most environments, a 1:10 performance:capacity ratio on FabricPool aggregates is extremely conservative while providing significant storage savings.
Note: Writes from the capacity tier to the performance tier are disabled if performance tier capacity is greater than 70%. If this occurs, blocks are accessed directly from the capacity tier.
Inactive Data Reporting
First available in ONTAP 9.4, inactive data reporting (IDR) is an excellent tool for determining the amount
of inactive (cold) data that can be tiered from an aggregate.
IDR uses a 31-day cooling period to determine what data is considered inactive.
Note: The amount of cold data tiered is dependent on the tiering policies set on the aggregate volumes. This amount might be different than the amount of cold data detected by IDR using a 31-day cooling period.
OnCommand System Manager
IDR is displayed on the Storage Tiers page in OnCommand System Manager.
Figure 10) Inactive data reporting.
ONTAP CLI
To display IDR using the ONTAP CLI, run the following command:
storage aggregate show-space -fields performance-tier-inactive-user-data, performance-tier-
inactive-user-data-percent
IDR is enabled by default on FabricPool aggregates. IDR can be enabled on non-FabricPool aggregates
using the ONTAP CLI.
ONTAP CLI
To enable IDR on a non-FabricPool aggregate, run the following command:
storage aggregate modify -aggregate <name> -is-inactive-data-reporting-enabled true
34 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
6.2 Connectivity
FabricPool read latency is a function of connectivity to the external capacity tier. LIFs using 10Gbps ports
provide adequate performance. NetApp recommends validating the latency and throughput of your
specific network environment to determine the impact it has on FabricPool performance.
Object Store Profiler
Starting in ONTAP 9.4, an object store profiler is available through the CLI that lets you test latency and
throughput performance of object stores before you attach them to FabricPool aggregates.
Note: The external capacity tier must be added to ONTAP before it can be used with the object store profiler.
Start the object store profiler.
storage aggregate object-store profiler start -object-store-name <name> -node <name>
(Advanced privilege level required.)
View the results.
storage aggregate object-store profiler show
Note: External capacity tiers do not provide performance similar to that found on the performance tier (typically GB per second).
Although FabricPool aggregates can easily provide SATA-like performance, they can also tolerate
latencies as high as 10 seconds and low throughputs for tiering solutions that do not need SATA-like
performance.
When using FabricPool in low-performance environments, minimum performance requirements for client
applications must continue to be met, and recovery time objectives (RTOs) should be adjusted
accordingly.
Amazon S3
Although AWS Direct Connect provides better performance and lower data transfer charges, it is not
required by FabricPool. Because performance is significantly better when using AWS Direct Connect,
doing so is the recommended best practice for FabricPool.
AWS Direct Connect network connections are available at 10Gbps connection speeds using cross-
connect. For more information about available bandwidth connections for Direct Connect, see the AWS
Direct Connect FAQs.
Microsoft Azure Blob Storage
Although ExpressRoute provides better performance and lower data transfer charges, it is not required by
FabricPool. Because performance is significantly better when using ExpressRoute, doing so is the
recommended best practice for FabricPool.
ExpressRoute network connections are available at 10Gbps connection speeds using cross-connect. For
more information about available bandwidth connections for ExpressRoute, see the ExpressRoute pricing
page.
Cloud Volumes for ONTAP
By default, network traffic in Cloud Volumes for ONTAP instances are generally limited to 2Gbps
(~250MBps). Starting in Cloud Volumes for ONTAP 9.4, SR-IOV instances (10Gbps) are supported.
35 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
StorageGRID Webscale
StorageGRID performance scales with the number of nodes in a system. For acceptable performance
targets, NetApp recommends using enough nodes to meet or exceed FabricPool connectivity
requirements.
SnapMirror Concurrency
Because concurrent SnapMirror and SnapVault replication operations share the network link to the
external capacity tier, initialization and RTO are dependent on the available bandwidth and latency to the
capacity tier. Performance degradation might occur if connectivity resources become saturated.
Proactive configuration of multiple LIFS can significantly decrease this type of network saturation.
Loss of Connectivity
If for any reason connectivity to the external capacity is lost, the FabricPool performance tier remains
online, but applications receive an error message. Cold blocks that exist exclusively on the capacity tier
are unavailable until it comes back online.
Error messages include:
• SMB STATUS_INTERNAL_ERROR
Client applications may or may not retry upon receiving this error. (It is client dependent.) The client does not have to remount.
• NFS v3: EJUKEBOX
v4: EDELAY
NFS client applications retry after five seconds. The NFS client hangs until connectivity is reestablished if it gets the same error after a retry.
• SAN UNRECOVERED READ ERROR/RECOMMEND REWRITE THE DATA
If the host is connected to the ONTAP LUN and the LUN is configured in a RAID set on the host (for example, Volume Manager), the host RAID subsystem might be able to recover the data from parity, and the data is rewritten to a new location. If the host is unable to recover this data, the application on the host might need to be restarted so that the read can be retried.
6.3 Capacity
The previously mentioned 1:10 performance:capacity ratio is conservative. FabricPool continues to tier
cold data to an external capacity tier up to a 1:20 performance:capacity ratio. For example, 16PB can be
tiered from an 800TB aggregate: in total, a 16.8PB FabricPool.
Volumes in a FabricPool aggregate cannot exceed 100TB regardless of on what tier the data is located.
After the 1:20 performance:capacity ratio has been reached, FabricPool stops tiering cold data to the
external capacity tier. If the performance tier reaches maximum capacity, cold data is read directly from
the external capacity tier without migrating to the performance tier.
A capacity warning is triggered when the capacity tier reaches 85% of the maximum capacity set by the
capacity-based license. Tiering to the capacity tier stops when the amount of data (used capacity) stored
on the third-party capacity tier reaches the licensed capacity. Additional data, including SnapMirror copies
to volumes using the backup tiering policy, cannot be tiered until the license capacity is increased.
Although tiering stops, data remains accessible from the capacity tier. Cold data remains on SSDs until
the licensed capacity is increased.
36 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Aggregate
Aggregate space utilization can be determined by using OnCommand System Manager or the ONTAP
CLI.
OnCommand System Manager
In OnCommand System Manager, FabricPool space utilization is displayed on the Storage Tiers tab.
Details include performance tier maximum capacity, used capacity, and external tier used capacity.
Figure 11) Storage tiers aggregate-level information.
Additionally, used capacity as a percentage of the licensed capacity is displayed at the bottom of the
Storage Tiers tab.
Figure 12) Licensed capacity.
ONTAP CLI
To view FabricPool aggregate space utilization details using the ONTAP CLI, run the following command:
storage aggregate object-store show-space
Example:
storage aggregate object-store show-space
License
Aggregate Object Store Name Provider Type Used Space Space Used%
-------------- ----------------- ------------- ------------- ------------
aggr1 aws_bucket AWS_S3 423.3GB 41%
1 entries were displayed.
Volume
FabricPool volume space utilization can be determined by using OnCommand System Manager or the
ONTAP CLI.
OnCommand Cloud Manager
In OnCommand Cloud Manager, S3 used capacity is displayed on the Volumes tab. Details include tier
(performance and capacity) allocated capacity and used capacity for both performance and capacity tiers.
37 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Figure 13) Volume capacity displayed in OnCommand Cloud Manager.
ONTAP CLI
To view FabricPool volume space utilization details using the ONTAP CLI:
volume show-footprint
Total, performance, and capacity tier (using the bucket name) footprints are displayed.
Vserver : svm_fabricpool
Volume : project_b
Feature Used Used%
-------------------------------- ---------- -----
Volume Data Footprint 16.84GB 1%
Footprint in Performance Tier 131.7MB 1%
Footprint in my-bucket 16.74GB 99%
Volume Guarantee 0B 0%
Flexible Volume Metadata 429.1MB 0%
Delayed Frees 27.60MB 0%
Total Footprint 17.29GB 1%
Available License Capacity
To view the capacity status of the FabricPool license using the ONTAP CLI, run the following command:
system license show-status
Example:
system license show-status
Status License Scope Detailed Status
--------- ------------------- --------- ----------------------
valid
NFS site -
CIFS site -
iSCSI site -
FCP site -
SnapRestore site -
SnapMirror site -
FlexClone site -
FabricPool cluster The system is using 423.3GB, and can use up to 10TB.
not-installed
SnapVault - -
SnapLock - -
SnapManagerSuite - -
SnapProtectApps - -
V_StorageAttach - -
Insight_Balance - -
OCShift - -
38 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
TPM - -
VE - -
DP_Optimized - -
not-applicable
Cloud - -
Select - -
20 entries were displayed.
To view the capacity status of the FabricPool license using OnCommand System Manager, complete the
following steps:
1. Click Configurations.
2. Click Licenses.
3. Click Details.
4. Click FabricPool License.
5. Current capacity is listed in the Current Capacity column.
Figure 14) License capacity.
39 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
6.4 NetApp Private Storage for AWS
NetApp Private Storage (NPS) for AWS meets or exceeds all FabricPool best practices. The NPS for
AWS solution is a high-performance cloud-connected storage architecture that allows enterprises to build
an agile cloud infrastructure that combines the scalability and flexibility of the AWS cloud with the control
and performance of NetApp storage.
NPS for AWS is typically deployed at one of the many AWS-approved Direct Connect partner colocation
data centers (for example, Equinix) and leverages AWS Direct Connect to provide a low-latency, highly
available, dedicated connection between NetApp storage and the AWS cloud.
40 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Where to Find Additional Information
To learn more about the information described in this document, refer to the following documents and/or
websites:
• Cluster Management Using OnCommand System Manager https://library.netapp.com/ecm/ecm_download_file/ECMLP2838139
• ONTAP 9 Disks and Aggregates Power Guide https://library.netapp.com/ecm/ecm_download_file/ECMLP2496263
• Commands: Manual Page Reference https://library.netapp.com/ecm/ecm_download_file/ECMLP2838138
• Deploying and Managing ONTAP Cloud Systems https://library.netapp.com/ecm/ecm_download_file/ECMLP2839261
• Updating and Administering Cloud Manager https://library.netapp.com/ecm/ecm_download_file/ECMLP2735479
• Storage Tiering with Cloud Volumes ONTAP http://clouddocs.netapp.com/occm/en/concept_storage.html
• StorageGRID Webscale Administrator Guide https://library.netapp.com/ecm/ecm_download_file/ECMLP2753104
• TR-4015: SnapMirror Configuration and Best Practices Guide http://www.netapp.com/us/media/tr-4015.pdf
• TR-4183: SnapVault Best Practices Guide https://www.netapp.com/us/media/tr-4183.pdf
• TR-4075: DataMotion for Volumes http://www.netapp.com/us/media/tr-4075.pdf
• TR-4133: NetApp Private Storage for Amazon Web Services (AWS) http://www.netapp.com/us/media/tr-4133.pdf
• TR-4695: Database Storage Tiering with FabricPool https://www.netapp.com/us/media/tr-4695.pdf
• NVA-0009: NetApp Private Storage for Cloud http://www.netapp.com/us/media/nva-0009.pdf
• ONTAP FabricPool Licensing Overview https://kb.netapp.com/support/s/article/ka21A0000008qb3QAA/ONTAP-FabricPool-FP-Licensing-Overview
41 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Version History
Version Date Document Version History
1.6 July 2018 John Lantz: Cloud ONTAP renamed to Cloud Volumes ONTAP. Added additional information regarding metadata.
1.5 June 2018 John Lantz: Support for tiering to Microsoft Azure Blob Storage, the Auto volume tiering policy, and io1 EBS volumes added to ONTAP Cloud. Writes from the capacity tier to the performance tier are disabled if performance tier capacity is greater than 70%.
1.4 May 2018 John Lantz: Updated for ONTAP 9.4.
Added Auto tiering policy, Microsoft Azure Blob support, inactive data reporting, and support for ONTAP Select Premium.
1.3 January 2018 John Lantz: Updated for ONTAP 9.3. Added ONTAP Cloud functionality, AWS GovCloud S3, and additional interoperability details (QoS, StorageGRID Webscale, etc.).
1.2 September 2017 John Lantz: Added details regarding connectivity requirements.
1.1 August 2017 John Lantz: Added details regarding intercluster LIF requirements.
1.0 June 2017 John Lantz: Initial commit.
Contact Us
Let us know how we can improve this technical report.
Contact us at [email protected].
Include TR-4598: FabricPool Best Practices in the subject line.
42 TR-4598: FabricPool Best Practices in ONTAP 9.4 © 2018 NetApp, Inc. All rights reserved. Public
Refer to the Interoperability Matrix Tool (IMT) on the NetApp Support site to validate that the exact product and feature versions described in this document are supported for your specific environment. The NetApp IMT defines the product components and versions that can be used to construct configurations that are supported by NetApp. Specific results depend on each customer’s installation in accordance with published specifications.
Copyright Information
Copyright © 2017–2018 NetApp, Inc. All rights reserved. Printed in the U.S. No part of this document covered by copyright may be reproduced in any form or by any means—graphic, electronic, or mechanical, including photocopying, recording, taping, or storage in an electronic retrieval system—without prior written permission of the copyright owner.
Software derived from copyrighted NetApp material is subject to the following license and disclaimer:
THIS SOFTWARE IS PROVIDED BY NETAPP “AS IS” AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WHICH ARE HEREBY DISCLAIMED. IN NO EVENT SHALL NETAPP BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
NetApp reserves the right to change any products described herein at any time, and without notice. NetApp assumes no responsibility or liability arising from the use of products described herein, except as expressly agreed to in writing by NetApp. The use or purchase of this product does not convey a license under any patent rights, trademark rights, or any other intellectual property rights of NetApp.
The product described in this manual may be protected by one or more U.S. patents, foreign patents, or pending applications.
RESTRICTED RIGHTS LEGEND: Use, duplication, or disclosure by the government is subject to restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.277-7103 (October 1988) and FAR 52-227-19 (June 1987).
Trademark Information
NETAPP, the NETAPP logo, and the marks listed at http://www.netapp.com/TM are trademarks of NetApp, Inc. Other company and product names may be trademarks of their respective owners.